Strategies used to defendpharmaceutical brands from

genericsDean C.H. Wilkie

Faculty of Pharmacy, University of Sydney, Sydney, Australia

Lester W. JohnsonMelbourne Business School, The University of Melbourne, Carlton,

Australia, and

Lesley WhiteFaculty of Pharmacy, University of Sydney, Sydney, Australia

Abstract

Purpose – This research aims to provide an empirical comparison of the results of three brands’marketing defence strategies used in advance of generic brands entering the market. By reviewing theeffectiveness of these strategies, this research looks to extend the research on marketing defencestrategies into the importance of anticipating competitor launches.

Design/methodology/approach – A data set containing 243 weeks of scanned sales for 21 genericbrands was used in a regression model aimed at measuring the effectiveness of each brand’s defencestrategies in deterring entry and limiting the market share of these generic brands.

Findings – The analysis shows that several marketing mix components were effective in limiting theimpact of generic brands. What was critical to each component’s success was ensuring that they wereimplemented before the launch of the generic brands.

Research limitations/implications – This research has the limitation of being confined to acategory of pharmaceutical allergy brands, which limits generalisation of the findings.

Practical implications – The managerial relevance of this research has two parts. First, it willencourage managers to move from implementing strategies in reaction to a competitor launch toimplementing strategies in advance of their entry. Second, it provides insights into the effectiveness ofseveral strategic options for brands facing the entry of generic brands.

Originality/value – This study brings together literature regarding entry deterrence and marketshare loss prevention to help highlight the importance of proactive marketing defence strategies inreducing both the number of entrants and the amount of market share lost. It uses a data set to providean empirical review of a range of marketing mix components used by pharmaceutical brands againstlow-price generic brands.

Keywords Brand management, Marketing strategy, Pharmaceutical products, Pricing, Generics,Competitive strategy, Premier brands

Paper type Research paper

IntroductionThe proliferation and market share growth of generic brands continues to be a threat toany pharmaceutical brand which has lost or about to lose the benefits of patentprotection. These pharmaceutical brands are the originators of a patented productformulation and have invested resources into developing and marketing the product’s

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0309-0566.htm

Strategies usedto defend brands

1195

Received 1 November 2009Revised 31 March 2010Accepted 23 June 2010

European Journal of MarketingVol. 46 No. 9, 2012

pp. 1195-1214q Emerald Group Publishing Limited

0309-0566DOI 10.1108/03090561211247883

point of difference to the competitors, with the objective of building brand preferencewithin the market (Reiffen and Ward, 2007; Scott Morton, 2000). However, the expiry ofa patent represents the opportunity for other manufacturers and distributors to sellbioequivalent versions, or generic copies, of the original branded product (ScottMorton, 2000). The majority of these generic copies are sold either under new brandnames or store brand labels and are positioned as equal in efficacy, therefore they canleverage some of the associations consumers have of the branded product, but with theadditional benefit of a significantly lower price (Caves et al., 1991; Berndt et al., 2003;Wiggins and Maness, 2004).

The entry of generic brands in any industry causes a shift in consumer attitudesand behaviour that ultimately change the market dynamics. By comparing theattitudes of consumers towards pharmaceutical generic brands and grocery genericbrands, several similarities become evident. In both industries it has been establishedthat the consumers make a trade off between quality and price when evaluating ageneric brand (Steiner, 2004; Sullivan et al., 1994). What makes this advantageous forthe generic brands is that majority of consumers in both industries believe that genericbrands are equal in quality to the branded products (Steiner, 2004; Sansgiry et al., 2005;Sullivan et al., 1994). In comparing price, both pharmaceutical and grocery genericbrands have been shown to offer similar price savings compared to the brandedproduct of between 20-30 per cent (Regan, 2008; Steiner, 2004). While there aresimilarities across industries in consumer acceptance of generic brands, a major pointof difference within the pharmaceutical industry is the effect of the patent expirationdate. In many other industries generic brands can enter at any time. Generic brands inthe pharmaceutical industry are restricted from entering until the patent expires, whichgives each generic brand an equal opportunity within the market. This opportunity cancreate intense competition which is highlighted by the fact that for everypharmaceutical brand that comes off patent, there is an average of five genericbrands launched (Regan, 2008). The high number of generic brands with perceivedequal quality to the branded product has led to generic brands being able to regularlyachieve a total market share in excess of 50 per cent (Regan, 2008; Aronsson et al.,2001). Due to the size of this threat, branded pharmaceutical products need to betterunderstand what marketing defence strategies are the most effective at limiting theimpact of generic brands.

The main purpose of a marketing defence strategy is to effectively align andallocate resources across a range of marketing mix components in order to defend abrand’s market sales or share from a new competitor entrant (Walker et al., 2003). Mostof the research into marketing defence strategies has either examined the strategiesused to only deter entry or the strategies used to limit market share losses in reaction toa competitor launch (Kuester et al., 1999). These two research streams ignore theopportunity some brands have to be proactive and implement marketing defencestrategies in advance of a competitor launch. This opportunity occurs regularly withinthe pharmaceutical industry where brands can anticipate competitors entering when apatent expires. Brands can also review the registration lists of governing bodies fornew competitor registrations as new entries are listed at least several months beforethe physical launch (Hollis, 2003; Reiffen and Ward, 2005).

The goal of this research is to merge the two research streams by examining theeffectiveness of marketing defence strategies implemented in anticipation of a

EJM46,9

1196

competitor launch. It reviews the effectiveness of these proactive strategies in deterringthe entry of competitors and limiting the market share of any new entrant. By doingthis, the research seeks to highlight and extend the focus of marketing defencestrategies to the benefits of implementing a marketing defence strategy in advance ofcompetitor launches.

The success of a marketing defence strategy relies on an understanding of theeffectiveness of each of the marketing mix components (Gatignon et al., 1997). Thispaper will investigate a range of components used in defending pharmaceutical brandsagainst the launch of generic brands. The marketing mix components includeadvertising, extending a brand’s product range through new products and lineextensions, modifying the price, and launching an own-brand generic product. Severalresearchers have provided theoretical and case study reviews of using these marketingmix components, however they have lacked an empirical analysis (Chandon, 2004;Barak and Wilson, 2003; Mehta and Mehta, 1997). Research with an empirical analysishas mainly focused on strategies for deterring entry (Kong and Seldon, 2004; ScottMorton, 2000; Ellison and Ellison, 2007) or the effect of only one of these marketing mixcomponents on the market share of generic brands (Huskamp et al., 2008; Reiffen andWard, 2007; Grabowski and Vernon, 1992). Our research is the first study to provide anempirical review of each of these marketing mix components within thepharmaceutical industry. It examines the marketing defence strategies of threeover-the-counter (OTC) brands of allergy products and describes how each brand’smarketing defence strategy impacts the entry and market share of the generic brands.

The data set used also provides further points of difference to previous research intopharmaceutical generic brands. First, previous research has used data that containseither prescription only brands or a composite of prescription and OTC brands.Analysing any data set that contains prescription products needs to consider theinfluence which a doctor plays in the brand choice. Using a data set containing onlyOTC brands removes the influence of the doctor and increases the role of marketingmix variables such as advertising, price and distribution as consumers are able to selfselect brands. Lastly, this research is the first to highlight that its data set of genericbrands contains store brands. Including store brands is important as they represent 20per cent of all retail purchases (Ailawadi et al., 2008), are growing at twice the rate ofbranded products (Sethuraman, 2009), and have several benefits compared to othergeneric brands (Sudhir and Talukdar, 2004). By using a data set where nearly half ofthe generic brands are store brands, it means that the results will be more robust andhave wider implications in developing a marketing defence strategy.

The results of this research provide significant insights into the effectiveness of arange of marketing mix components used to defend a branded product. It alsohighlights the benefits of implementing these components before a competitor enters.More importantly, the results and benefits of implementing strategies beforecompetitors enter should not be viewed as being restricted to the pharmaceuticalindustry where it may be easier to anticipate a competitor entry. In many industriesoutside the pharmaceutical industry brands become aware of new competitors,branded or generic, before they launch which gives a brand the opportunity toimplement marketing defence strategies in advance. It is hoped that this study willencourage further research into this area of proactive defence strategies across a rangeof brands and industries.

Strategies usedto defend brands

1197

This paper is organised as follows. The next section reviews the relevant literatureon each of the marketing mix components used in the marketing defence strategies.The third section provides a description of the data and the models used to test thehypotheses. The results are discussed next, and finally the conclusion, limitations andproposals for future research.

Marketing mix componentsConsumer advertisingConsumer advertising has several advantages in defending a brand from a newcompetitor entrant. It places the brand top of mind with consumers (Sutherland andSylvester, 2000), it assists in consumer learning and forming of preferences (Carpenterand Nakamoto, 1989), it reinforces brand loyalties which helps reduce a consumer’sprice sensitivity (Frank and Salkever, 1992), it can be used to encourage purchases(stocking up) before the launch of a competitor, and finally it can create a barrier toentry (Karakaya and Stahl, 1989; Thomas, 1999).

However, the effectiveness of this approach has been questioned due to theconflicting results in the literature. Advertising can be a barrier to entry if it raises thecost of entry for a competitor (Thomas, 1999). However because the cost of entry for ageneric brand is relatively low, the use of advertising as a barrier to entry has beenshown not to deter entry (Scott Morton, 2000). Advertising may actually encourage ageneric brand to enter if the advertising helps increase the category growth and/or size.This is because the profit expectations of generic brands will increase with a largercategory size (Scott Morton, 2000). These conflicting results indicate that a brand’sadvertising can increase or decrease the number of generic brand entrants:

H1. A brand’s advertising has an impact on the number of generic brand entrants.

The impact of using advertising has also been influenced by the behaviour of brandsbefore the launch of a generic competitor. Research has shown that many brandsreduce their advertising spends prior to, and after the launch of generic brands(Huskamp et al., 2008; Iizuka, 2004; Berndt et al., 2003; Ellison and Ellison, 2007). Caveset al. (1991) found that brands reduce their advertising spend up to two years prior tothe launch of the generic brands. The spend fell an additional 20 per cent with the entryof the first generic brand, 40 per cent when the number of entrants is five and anadditional 20 per cent when the number of generic brand hits ten. If a brand reduces itsadvertising spend before and after the launch of generic brands, then its effectivenessas a component of a marketing defence strategy diminishes (Thomas, 1999).

The drop in advertising spend reflects a profit maximising strategy and may nothave anything to do with the effectiveness of advertising as a component of amarketing defence strategy. Gatignon et al. (1997) provides evidence that marketingexecutives believe advertising is positively associated with a successful marketingdefence strategy. With varying evidence as to the success of using advertising, it isimportant to establish how advertising impacts the share of generic brands:

H2. A brand’s advertising impacts the market share of generic brands.

Extending the product range through innovations and line extensionsExtending the brand’s product range through launching innovative new products andline extensions is a strategy aimed at shifting consumer preferences towards the new

EJM46,9

1198

products in the range before the introduction of a competitor entrant (Reiffen andWard, 2007). Consumer preferences are shifted towards the new product as it offers apoint of difference to the brand’s current products that better suits their needs. This inturn reduces the amount of profitable positions for a competitor entrant to target(Thomas, 1999).

Shifting consumer preferences through the introduction of innovative products hasbeen shown to be a more effective strategy in protecting market share fromcompetitors in comparison to a low price strategy (Kuester et al., 1999; Gatignon et al.,1997; Shankar et al., 1998). There are other ways in which innovation can help a branddefend its share from new competitors. Introducing new technology may helpdiscourage competitors from entering as it signals that the brand is investing resourcesinto growing and the innovation makes existing technology outdated (Hauser et al.,2006). Innovation has also been shown to increase the sales of existing products byhelping reposition the brand in the mind of consumers through the tangible andintangible attributes of the innovation (Kapferer, 2008). Due to the ability of aninnovative new product to shift consumer preferences, discourage competitors andincrease the sales of existing products, it is anticipated that the introduction of aninnovative product will reduce the market share of the generic brands:

H3. The introduction of an innovative product will reduce the market share ofgeneric brands.

As part of a marketing defence strategy, the introduction of a new innovative productmeans that brands need to consider whether to use the existing brand name or a newbrand name (Kerin et al., 1996). Within the pharmaceutical industry this decision canbe influenced by Government regulations aimed at removing the safety risk ofconsumer confusion between different active ingredient combinations. As a result ofthis regulation, many brands are forced to either use a hybrid name containing onlypart of the parent brand name or to use a new brand name.

The benefit of using an existing or hybrid brand name is that it costs less tointroduce and leverages the parent brand’s image which helps increase brand trial(Kerin et al., 1996). Kerin et al. (1996) also showed that brand extensions with the samebrand name are more responsive to changes in marketing mix variables than newbrand names:

H4. The introduction of a hybrid brand name innovation will have more effect onthe generic brands market share than a new brand name innovation.

Introducing line extensions is a common strategy used to defend a brand’s marketshare from competitors (Bayus and Putsis, 1999). A line extension is defined as when abrand introduces a new product, or stock keeping unit, that differs from the currentrange of products in a minor way such as format, flavour or pack size (Kapferer, 2008).There are several ways in which line extensions can be an effective component of amarketing defence strategy. Introducing new line extensions increases the costs for acompetitor to replicate the range, and if the range is not replicated, the new entrant willlimit its ability to switch consumers (Ellison and Ellison, 2007). Increasing the numberof stock keeping units within a brand’s range before the launch of a new competitorwill also help protect shelf space which has been shown to reduce the market share ofthe new entrant (Kalyanaram and Wittink, 1994):

Strategies usedto defend brands

1199

H5. The more stock keeping units a brand has in its range, compared to thenumber of stock keeping units within the generic brand’s range, the greaterthe reduction in the market share of the generic brands.

Modifying the priceThe entry of generic brands can see the average price within a category fall by over 60per cent as each generic brand competes to establish a low price position (Caves et al.,1991; Berndt et al., 2003; Wiggins and Maness, 2004). This significant difference inprice has focused much of the research into the relationship between branded productsand generic brands on price. When a low price generic brand enters the market itrepresents a competitive threat that brands must address, however there is muchdebate as to whether price is the most effective marketing mix component to use. ScottMorton (2000) suggested that if a brand lowers its price before the entry of genericbrands it can act as a barrier and deter entry. The reduction in price acts as a signal topossible entrants that the lower price means that they should expect lower profits.However brands are driven by profit maximising objectives and reducing the price tothe levels needed to deter entry are yet to be observed (Grabowski and Vernon, 1992;Frank and Salkever, 1992). Caves et al. (1991) showed that the average price for abranded product fell by only 2 per cent with the launch of the first generic brand.

Since brands demonstrate a profit maximising behaviour, Frank and Salkever(1992) and Regan (2008) suggest that the introduction of a generic brand segments themarket into a price sensitive segment and a brand loyal segment. As part of the profitmaximising behaviour, a brand should then sacrifice the price sensitive segment infavour of the brand loyal segment which will allow the brand to maintain higher pricesand maximise profits. Brands may even consider increasing the price as the brandloyal segment is more price insensitive (Frank and Salkever, 1992; Regan, 2008). Afurther implication is that after the market has been segmented, generic brandscompete amongst themselves and not against the branded product (Reiffen and Ward,2005; Regan, 2008). Therefore a change in the pricing strategy of the branded productshould not have an impact on the market share of generic brands:

H6. Increases (decreases) in a brand’s price will not increase (decrease) the marketshare of generic brands.

Launching an own-brand genericThere is a growing popularity for brands to launch their own low price generic versionof the branded product under a new brand name. In Australia, Hollis (2003) estimatedthat 25 per cent of generic brands in this market are own-brand generics. Theseown-brand generics may benefit from the resources of the branded product (i.e. lowercosts of manufacture, larger sales team) but the success of launching an own-brandgeneric depends more on being the first generic to market (Grabowski and Vernon,1992). This can be easily achieved as an own-brand generic is not restricted by patentprotection laws and has regulatory advantages over other generic brands (Reiffen andWard, 2007).

There are several strategic reasons why a brand should launch an own-brandgeneric and gain the first to market advantage. This includes the impact it has onreducing the market share of the other generic brands, helping control the prices of the

EJM46,9

1200

other generic brands and reducing the number of generic brand entrants (Hollis, 2002;Scott Morton, 2000; Reiffen and Ward, 2007).

Early research into the advantage of being first to market established that a brandcan create a sustainable market share advantage through influencing consumerpreferences and taking ownership of a desirable position (Carpenter and Nakamoto,1989). Like first to market brands, the first generic brand to launch helps establishconsumer preference and own the low price position. Hollis (2002) highlighted that thefirst to market generic brand receives higher preferences with pharmacists as well.This is because it is easier to switch a consumer to the first generic brand because itoffers a significant cost saving whereas following generic brands offer less of a costsaving. Later generic entrants also face barriers in gaining distribution as pharmacistsseek to minimise inventory costs and have limited shelf space (Hollis, 2002):

H7. The first generic brand to market gains a market share advantage over theother generic brands.

The first to market advantage is enhanced by the amount of time between the launchand that of the next competitor entrant (Vakratas et al., 2003). The more time betweenentry and the next competitor allows a brand to build distribution and preferences withconsumers and pharmacists. Grabowski and Vernon (1992) showed, within their studyof 18 brands that came off patent, that in 50 per cent of cases the first to market genericbrand was the market leading generic brand. However if the first to market generic hadat least three months between itself and the next entrant, in all cases it was the marketleading generic. This finding that the time between generic launches impacts marketshare has received further support from Hollis (2002) and Kamien and Zang (1999).Since a brand can control the launch timings of the own-brand generic, capitalising onthis advantage is essential to limiting the market share of the other generic brands:

H8. The more time between the launch of a generic brand and the next genericcompetitor, the greater the reduction in market share of the next generic brandentrant.

Being the first to market with an own-brand generic has further benefits in defending abrand’s market share. The launch of an own-brand generic before patent expiry signalsto other potential entrants that their market share and potential profits will be reduced(Reiffen and Ward, 2007). The lower market share potential reduces the number ofgeneric entrants due to the lower potential profits (Scott Morton, 2000; Reiffen andWard, 2007). Reiffen and Ward (2007) showed that the launch of an own-brand genericreduces the number of entrants by approximately two:

H9. The launch of an own-brand generic will reduce the number of generic brandentrants.

H10. The launch of an own-brand generic will reduce the market share of genericbrands.

Data and empirical modelThe data consist of aggregated national sales from the Australian market and weresupplied by Aztec Synovate data. They contain 243 weeks of scanned consumer saleswithin the allergy category of OTC treatments. The category averages annual sales of

Strategies usedto defend brands

1201

$92 million Australian dollars with the three brands within the study competingintensely for share as each brand is registered for treating seasonal and perennial allergicrhinitis. Each of the three brands lost patent protection within two years of each otherwhich has led to a proliferation of generic brands. In total there are 21 brands of genericswith the number targeting each brand varying from four to 12. The data are ideallysuited for a comprehensive review of the marketing defensive strategies used againstgeneric brands as each brand implemented varying strategies. Table I provides asummary of marketing mix components and the differing strategies implemented byeach brand.

In reviewing each of the marketing mix components used in this case study, eachbrand predominately used television as its choice for advertising. After the launch ofthe generic brands, brands 1 and 2 increased their advertising spend to help supportthe innovative new products. Brand 3 demonstrated a profit maximising behaviourand reduced the average amount it spent on advertising.

Brands 1 and 2 have introduced new innovative products with an improvedformulation aimed at providing a significant point of difference to the generic brandsand gaining a new patent protected molecule within the category. These two brandshave also established a larger pack size before the launch of the generic brands. Thisprovides an ideal example of a line extension as the larger pack sizes target a specificsegment of frequent sufferers of allergies. It offers the target user increased value pertablet compared to the other products within the brand’s range and also reduces theprice per tablet difference to the generic brands. Increasing the pack size also serves thepurpose of reducing the number of purchasing occasions that could lead to theconsumer switching to a generic brand.

Brands 2 and 3 have introduced an own-brand generic as part of their marketingdefence strategy. Both brands introduced the own-brand generics nine months beforethe patent expired to ensure these generic brands gain a first to market advantage.Brand 3 also launched a second own-brand generic 34 months after its patent expired.

Table I also highlights that brand 3 was the only brand to use price as a componentof the marketing defence strategy. It reduced its price by more than 11 per cent beforethe introduction of the first generic brand. By reducing the price before the launch ofthe generic brands it ensured the new price was established, could act as a signal to thegeneric brands, and had the maximum opportunity to succeed. Since the initialreduction, Brand 3 has steadily increased its price but it still remains below the pricelevel before the implementation of the defence strategy. The other two brands haveincreased their price annually with price rises of between 3-5 per cent.

Marketing variables and definitionsThere are ten explanatory variables to be used in the model developed for this study.Aztec Synovate data provided each of these measures. Price is the scanned price at theretail level which allows the model to determine the impact of a low price generic brandon consumer choice. Distribution is the percentage of stores selling the generic brandweighted by the pharmacy volume to the total market volume. Advertising is theamount of expenditure for the branded product and any advertising expenditure onnew innovative products, for that week. Advertising within the allergy category isprimarily directed towards consumers as OTC medicines allow for consumer selfselection. The number of stock keeping units in the product range is the number of

EJM46,9

1202

Bra

nd

1B

ran

d2

Bra

nd

3

All

erg

yca

teg

ory

aver

age

Ind

ust

ryav

erag

e

Summary

ofbranded

products

Av

erag

ep

rice

bef

ore

gen

eric

bra

nd

s($

)18

.36

17.9

616

.96

Av

erag

em

ark

etsh

are

bef

ore

gen

eric

bra

nd

s(%

)18

.538

.443

.1C

han

ges

top

rice

stra

teg

yIn

crea

seIn

crea

seR

edu

ced

Ow

n-b

ran

dg

ener

icN

oY

esY

es(2

bra

nd

s)In

trod

uce

dan

inn

ovat

ive

pro

du

ctN

ewb

ran

dn

ame

Hy

bri

db

ran

dn

ame

–In

trod

uce

da

lin

eex

ten

sion

Yes

Yes

No

Av

erag

ey

earl

yad

ver

tisi

ng

in$0

00s

(bef

ore

gen

eric

bra

nd

s)1,

370

650

1,75

0A

ver

age

yea

rly

adv

erti

sin

gin

$000

s(a

fter

gen

eric

bra

nd

s)1,

782

4,72

21,

260

Summary

ofgenericbrands

Av

erag

esh

are

ofac

tiv

ein

gre

die

nt

(%)

18.2

24.2

44.7

37.6

58.5

Av

erag

ed

istr

ibu

tion

(%)

13.1

21.5

24.3

19.6

NA

Av

erag

eg

ener

icp

rice

($)

13.9

616

.17

15.3

115

.15

NA

Nu

mb

erof

entr

ants

125

47

4.7

Dis

cou

nt

tob

ran

d(%

)27

.116

.213

.218

.822

.3N

um

ber

ofst

ore

bra

nd

s4

33

Nu

mb

erof

ind

epen

den

tb

ran

ds

82

1D

ate

ofp

aten

tex

pir

yJu

ly20

07Ju

ne

2006

May

2005

Table I.Case study summary

Strategies usedto defend brands

1203

individual product units for that brand. The order-of-entry and time between entries isdetermined through the first scanned date within the data set. The remaining variablesare used to measure the impact of each of the branded products strategies. Thesevariables include the launch of an own-brand generic, the introduction of a hybridbrand innovation and a new brand innovation.

Statistical modelThe analysis uses a log model to explain variations in the market share of genericbrands as a result of the marketing defence strategies of the branded products. A logmodel was selected as it allows for a comparison across the branded products as eachparameter represents variations in market share as a percentage. Using percentageshelps eliminate cross brand differences in marketing variables such as market shareand price (Kalyanaram, 2008). An example of a cross brand difference can be seen inTable I where brand 1 has less than half of the market share of the other two brands.

The dependent variable within the model is the ratio of the generic brand’s marketshare to the share of the branded product of the same active ingredient. A ratio wasselected as it eliminates the influence of other brands on the relationship between thegeneric brand and the branded product (i.e. if generic brand X reduces its price, it maygain several share points. However this share may come from switching users fromother generic brands where the objective is to measure if the changes in generic brandX’s price resulted in a change in the market share of the branded product). The marketshare of the branded product also contains the market share of the own-brand genericand the market share of any new innovative products. This assumption was madebecause these products were used by the branded product to help reduce the impact ofthe generic brands and as a result cannibalised some of the sales of the brandedproduct. If the model did not include the sales of these products, the market share of thebranded product would have a greater decrease at the same time as the generic brandsshare increases which may create misleading results.

A ratio of the generic brand to the branded product was also used for the price andstock keeping unit variables. This is because there is a two way relationship betweenthe two, as either brand can implement strategies using these variables which mayaffect the other brand’s share (i.e. if either brand was to reduce its price it could reducethe share of the other brand). The model takes the log of the generic brand’sdistribution as there is no evident relationship that changes to either the generic brandsor the branded products distribution impacts the others’ share. In addition each of thebranded products has maintained close to full distribution whereas the generic brandshave an average distribution of less than 20 per cent. For these reasons, a log modelwill determine how much market share a generic brand gains from an increase indistribution. With advertising, only the branded products were advertised, thereforethe model can only measure how changes to the branded product’s advertisingexpenditure impact the share of the generic brands. The model uses actual advertisingspend instead of a logarithm as logarithms of large numbers tend to create an artificialeffect of diminishing returns which underestimates the effects of advertising.

To measure if the first generic brand to market gains a market share advantage overthe other generic brands, two variables are included in the model. The first measures theimpact of the order-of-entry and the second measures the impact of time. A genericbrand’s order-of-entry is determined by its entry within the total category as opposed to

EJM46,9

1204

within the active ingredient group. This is based on the fact that these generic brandscan be interchangeable despite the different actives. A log model is used to measure theeffect of the order-of-entry as it assumes that the impact the first generic brand has on thebranded product will be more than the subsequent entrants (Kalyanaram and Wittink,1994). This assumption is consistent with the order-of-entry generalisation thatconsumer learning and preferences as well as market shares follow the sequential orderof entry (Carpenter and Nakamoto, 1989). Time is measured in terms of the number ofmonths between the last entrant and the next to enter. The last parts of the equation arethe dummy variables used to measure the effects of the own-brand generic, an innovativehybrid or new-brand name product. The formal equation is:

lnðSijtÞ ¼ aþ b: lnðPitÞ þ d: lnðDitÞ þ l:ðAijtÞ þ g: lnðRijtÞ þ t: lnðEijÞ

þ 1: lnðTijÞ þ 6ðOGiÞ þ fðOI iÞ þ rðNiÞ þ cðCompÞ þ eijt ð1Þ

where:

Ij is the number of brands with the same active ingredient j.

Tj is the number of time periods for active ingredient j.

J is the number of active ingredients.

Sijt is the market share ratio of the ith generic brand’s sales to the brand allergyproduct for active ingredient j in period t.

Pit is the ratio of the ith generic brand’s price to the branded product for activeingredient j in period t.

Dit is the ith generic’s total distribution in period t.

Aijt is the branded product’s advertising for active ingredient j in period t.

Rijt is the ratio of the ith generic brand’s range of stock keeping units to thepatent brand’s range of stock keeping units for active ingredient j in period t.

Eij is the order of entry for generic brand i within the allergy category.

Tij is the time (in months) between the entry of generic brand i and genericbrand i 2 1 within the category for active ingredient j.

OGj is a dummy variable used to represent the presence of an own-brand genericfor active ingredient j.

OIj is a dummy variable used to represent the presence of an own-brandinnovation for active ingredient j.

NIj is a dummy variable used to represent the presence of a new brandinnovation for active ingredient j.

Comp is a dummy variable used to account for company specific variables.

a is the constant term for the model.

eijt is the error term for the model.

Strategies usedto defend brands

1205

Empirical resultsTable II gives the parameter values for model 1 with each variable being significant atthe 5 per cent level with the exception of price ( p ¼ 0.318). The parameter values forthe own-brand generic and the new brand name innovation both have a positive valuewhich is different to the hypothesised but in both cases it is significant. A randomeffects model was used and tested with the Hausman test ( p ¼ 0.80) and was found tobe consistent and efficient (Hausman and Taylor, 1981). In addition the model used arobust estimator of the variance estimates to control for heteroskedasticity (Baum,2006). The R 2 value of 0.87 indicates that the model provides a strong statistical fit.

The following discussion of the results demonstrates the impact of each marketingmix component in restricting the market share potential of the generic brands. Toexplain the calculations, each parameter value estimates the impact of the componenton the average market share of the generic brands. To find the amount, the value of theparameter is converted from a logarithm into a normal value, and transferred into apercentage.

The first hypothesis (H1) predicts that a brand’s advertising will impact the numberof generic brand entrants. Table I reports the level of advertising expenditure for eachbrand and the number of generic brand entrants. The results provide no evidence thatthe level of advertising expenditure, before or after the launch of generic brands,impacts the number of entrants. Before the launch of the generic brands, brand 2 andbrand 3 had the highest and lowest levels of advertising expenditure respectively.After the launch of the generic brands the roles were reversed with brand 3 having thehighest and brand 2 having the lowest levels of advertising expenditure and yet bothbrands had a similar number of generic entrants (four and five entrants respectively).This number of generic brands was significantly lower than the 11 generic brandentrants that targeted brand 1, which had maintained a high level of advertisingexpenditure.

H2 predicted that the branded product’s advertising will impact the market share ofgeneric brands. The results in Table II show that a brand’s advertising does impact themarket share of generic brands ( p , 0.01). Using the average advertising expenditurefigures from Table I, each of the branded product has reduced the impact of genericbrands by an average of 22 per cent (brand 1 ¼ 16.3 per cent, brand 2 ¼ 37.6 per cent,

Explanatory variable Parameter value t-statistic

Log price ratio 20.06 21.03Log distribution 0.94 27.34Log stock keeping unit ratio 0.34 6.31Advertising 20.000145 22.54Order of entry 20.29 25.36Time between entry 20.07 22.79Own-brand generic 0.10 5.00Hybrid brand innovation 20.58 29.51New brand name innovation 0.13 5.24

Notes: Company specific variables not included as there are 14 companies; R 2 value ¼ 0.8687; no. ofobservations 2820; no. of groups 21

Table II.Empirical Results formarket share model

EJM46,9

1206

brand 3 ¼ 11.8 per cent). While advertising may not reduce the number of genericbrand entrants, the results provide evidence that it does help in defending a brand fromgeneric brands.

There were two hypotheses examining the impact of introducing an innovative newproduct as a component of a marketing defence strategy. The first (H3) predicted thatthe introduction of an innovative product would reduce the market share of genericbrands. The model estimated the impact of innovative products on the market share ofgeneric brands by splitting innovation into the two types, hybrid brand nameinnovation and a new brand name innovation. The parameter values for both types ofinnovative products are statistically significant ( p , 0.01) however one of the types ofinnovative products has a positive sign. Therefore H3 is partially supported by theresults as only one of the two types of innovative products has a positive impact on themarket share of generic brands.

H4 compared the effectiveness of a hybrid brand name innovation to a new brandname innovation in reducing the market share of generic brands. The results indicatethat introducing a hybrid brand name innovation limits the average market share ofthe generic brands by 44.0 per cent. In addition to this result, an order-of-entryadvantage was shown to exist within the category. This meant that introducing ahybrid innovation before the generic brands entered reduced the average market shareof the generic brands by a further 5.5 per cent.

The use of new brand name innovation increases the market share of genericsbrands. This effect may be explained by several factors, including the small samplesize of a single brand, the potential for consumer confusion in misattributing the newbrand name to a new generic brand with a similar name. Another major factor could bethat new brand name product was launched 14 months after the branded allergy patentexpired. In that time all the 11 generic brands for that active ingredient had launchedand may have established a level of consumer preference. This delay in launchingprevented the new brand name innovation from shifting consumer preferences beforethe generic brands entered or signalled the innovation to potential deter entrants.

H5 predicts that the more stock keeping units the branded-product has within itsrange compared to the amount of stock keeping units within the generic brands range,the greater the reduction in market share of the generic brands. The results fromTable II support this hypothesis. The parameter value of 0.34 is significant ( p , 0.01)and has the hypothesised sign. The impact of having an additional stock keeping unitcan be illustrated with the following example. One of the branded-products increasedits range from two stock keeping units to three stock keeping units by launching alarger pack size. The increase from two to three stock keeping units restricted theaverage market share of the generic brands by 12.8 per cent.

H6 states that an increase or decrease in the branded product’s price will not impactthe market share of generic brands. The results in Table II show that the parametervalue for changes in the pricing strategy of the brands is insignificant ( p ¼ 0.30). Thisresult supports H6 and the current evidence that pricing changes of the brandedproduct do not impact the market share of generic brands (Frank and Salkever, 1992;Regan, 2008).

H7 through to H10 examine the impact of launching an own-brand generic as partof a brand’s marketing defence strategy. The first hypothesis (H7) predicted that thefirst generic brand to launch gains a market share advantage over the other generic

Strategies usedto defend brands

1207

brands due to an order-of-entry effect within the category. The results in Table IIsupport this hypothesis as the order-of-entry parameter value has the hypothesisedsign and is significant ( p , 0.01). In addition to this advantage, the time between thelaunch of a generic brand and the next entrant was predicted to reduce the marketshare of the next entrant (H8). The results in Table II also show that the time betweenentries has a significant impact on the market share of the next entrant ( p , 0.01). Thefollowing example illustrates the size of the first to market advantage and the impact ofthe amount of time between entries. Brand 2 introduced its own-brand generic ninemonths before the next competitor. Using the two values from Table II and the brand’sorder of entry number (7 and 8 respectively), the next competitor’s market share wasrestricted by 19.5 per cent.

H9 predicted that an own-brand generic reduced the number of generic entrants.Table I provides support that the introduction of an own-brand generic may reduce thenumber of generic brand entrants. The one brand that did not have an own-brandgeneric had 11 generic brand entrants compared to the other two brands which bothhad five generic entrants. This finding is in line with the previous research of Reiffenand Ward (2007) which showed that an own-brand generic strategy reduced theaverage number of generic brand entrants by approximately two.

A major objective in launching an own-brand generic is to reduce the market shareof generic brands (H10). The results in Table II show that the introduction of anown-brand generic does not reduce the market share of generic brands ( p , 0.01). Theresults show that when an own-brand generic was launched, the generic brandsachieved greater market share compared to the brand 1 where no own-brand genericwas launched. The reason for this result may be caused by the number of and types ofgeneric brand entrants. In Table I the summary statistics show that brands 2 and 3have fewer entrants than brand 1, and of those entrants targeting brands 2 and 3, ahigher percentage is store brands. Store brands may be stronger competitors comparedto independent generics as store brand managers have the ability to implement moremarketing strategies that help in switching consumers from the branded products tothe store brand (i.e. these managers are responsible for determining the retail price,product ranging, and distribution for all brands within the category). In addition thehigher number of independent generic entrants targeting brand 1 may be the result of astrategic decision to avoid competing against an own-brand generic. The independentgeneric brands may be smaller with fewer resources and as a result more likely tounderperform compared to the average. Whereas the stronger store brands that arecompeting against brands 2 and 3 are more likely to perform above average. Thereforethis association of higher generic brand share, as a result of a branded productlaunching an own-brand generic, may be the result of a correlation with the type ofgeneric brand competitor more so than the effectiveness of an own-brand generic.

Launching an own-brand generic has been shown to reduce the number of entrantsand to gain an advantage through being the first generic brand in the market (i.e. it wasshown in H7 and H8 to reduce the share of the next entrant by 19.5 per cent), howeverH10 showed that an own-brand generic increased the market share of generic brandentrants. To help illustrate the net effect of launching an own-brand generic (H7, H8and H10) on the market share of the next generic brand to enter, brand 2’s own-brandgeneric restricted the market share of the next entrant by 10.2 per cent.

EJM46,9

1208

ConclusionThe objective of this research was to extend the literature into marketing defencestrategies by providing an empirical analysis of a range of marketing strategiesdeveloped in advance of a competitor entering. The learnings add to several researchstreams by highlighting the effectiveness of individual marketing mix components in amarketing defence strategy aimed at deterring potential entrants and limiting themarket share of any new competitors. It also highlights the importance ofimplementing these marketing mix components in advance of a competitor enteringthe market.

The research used a data set from a category of OTC products where three brandsof allergy treatments had lost patent protection and experienced a proliferation ofgeneric brands. Each of the three brands used a combination of several marketing mixcomponents such as advertising, introducing innovative new products and lineextensions, modifying the price, and launching an own-brand generic to form amarketing defence strategy. For a manager who wants to understand the effectivenessof these marketing mix components, launching an innovative new product under ahybrid brand name was the most effective strategy. If an innovative new product isunavailable then a manager should look to launch new products in different pack sizesand formats. Under both scenarios the benefit of being proactive and launching newproducts before the launch of the generic brands helped in shifting and reinforcingconsumer preferences and protecting the brand from losing market share to the genericbrands.

As part of a marketing defence strategy, managers should also consider advertisingas it was also shown to be effective in limiting the market share of the generic brands.There may also be additional benefits associated with advertising that this researchdid not measure, but that should be considered. First, the effectiveness of advertisingneeds to be measured against both generic brands and other branded competitors. Abrand’s advertising may be more effective in switching users from other brandedproducts than from generic brands due to the segmentation of the consumers into pricesensitive and brand loyal segments. An additional benefit may come from advertisingan innovative product. Advertising will help increase awareness of the new productand its attributes. As Kapferer (2008) highlighted, when a consumer learns about theattributes of an innovative product, it helps to reposition the entire brand which has ahalo effect on the sales of existing products.

The use of price as a component of a marketing defence strategy was shown to nothave an impact on the market share of the generic brands. This result is in line with theprevious research (Frank and Salkever, 1992; Regan, 2008), however this paper differsin its conclusions. Frank and Salkever (1992) and Regan (2008) suggest that a brandedproduct should increase its price as a result of the segmentation of the market into pricesensitive and brand loyal segments in order to maximise profits. As with advertising,our conclusion is that a manager needs to first understand how a change in its priceaffects the market share of other branded products. A price reduction may switch moreusers from other branded competitors than it will from generic brands with pricesensitive users.

The other two components of launching an own-brand generic or a new brand nameinnovation were also shown to not be effective in reducing the share of the genericbrands, however the influence of time and signalling to other competitors needs to be

Strategies usedto defend brands

1209

considered in evaluating this result. Time and the order in which a brand enters thiscategory were shown to impact the market share of generic brands. By launching anown-brand generic before the generic brands entered, it signalled to potential entrantsthat their market share and profits will be reduced, which may have helped to detercompetitors from entering, and secured an order-of-entry advantage. Because of thesetwo factors, the overall result was that the launching an own-brand generic did in factreduce the market share of the generic brands. A similar result would have beenexpected if the new brand name innovation launched before the generic brands, insteadit launched 14 months later and lost the opportunity to gain an order-of-entryadvantage and to signal the innovation to deter competitors from entering. These twoexamples further highlight the critical success factor of time in developing a marketingdefence strategy.

The existence of an order-of-entry advantage provides an important categoryinsight. It demonstrates that the generic brands that enter when the first patentexpires, achieve more share than generic brands that enter when later patents expire.This implies that the first branded product to lose patent protection will lose moremarket share. This however may be the result of price sensitive consumers beingwilling to switch from not just the branded product which the generic brands havereplicated, but from other branded products with a different active. Therefore if thefirst generic brands to enter are able to switch consumers from all brands, then allbranded products should consider implementing their marketing defence strategiesbefore the first generic brand for the category enters. For example a branded productmay want to introduce an own-brand generic not only before its own patent expires,but before the first patent in the category expires, in order to ensure it achieves a largershare of the price sensitive segment of consumers.

There are further implications that become evident as an indirect result of thisresearch. First, the results highlighted that distribution was a significant variable ingeneric brands gaining share. The branded products in this study tried to reduce thegeneric brands’ distribution by launching new products that created barriers togaining distribution and shelf space. Therefore managers should consider additionalstrategies which can limit the distribution of the generic brands. Conversely, managersof generic brands should actively pursue strategies that grow distribution andultimately market share. Store brands have the benefit of being able to gain fulldistribution within their retail outlets and be positioned next to the branded product onshelf. The use of a data set with nearly half the generic brands being low price storebrands, showed that the use of these marketing mix components were successfully indefending a brand from this increased threat.

Limitations and further researchThe limitation of this research is that it was confined to a single category of OTCallergy which limits the depth of investigation and the generalisation of findings. Itwas highlighted that some of the results may have been influenced by the number ofexamples within the data (e.g. there is only one example of each type of innovative newproducts). In order to overcome this limitation it is hoped that this study will lead tofurther research into this area with a more robust number of industries and brands.

Within this research several areas of further investigation become apparent, such asthe need for a greater understanding of the relationship between a brand’s advertising

EJM46,9

1210

or pricing with the market share of the other branded products (i.e. does a brand’sadvertising help switch more users from other branded products than from genericbrands?). Understanding this relationship would provide a more comprehensive view ofthe impact of a brand’s defence strategy on its market share. Another area of proposedresearch is the effect of brands “signalling” to other competitors their strategic directionthrough implementing strategies in advance of a competitor launching. Within the studyit was highlighted that the brand that did not signal any new product launches had morecompetitors and was at a market share disadvantage compared to the other two brands.More evidence is needed into whether “signalling” does reduce the number ofcompetitors and change the type of competitor. There are also some strategicconsiderations that need to be understood about launching an own-brand generic. Forexample does an own-brand generic increase the cannibalisation of the brand by havingshared resources? Does an own-brand generic increase the ability of pharmacists toswitch consumers away from the branded product? Does it increase the qualityperceptions consumers and pharmacists have the own-brand generic?

This research examined the marketing defence strategies of three brands regardingentry deterrence and limitation of the market share of generic brands. However theimpact of each strategy on maximising profit needs to be included in the developmentof an overall defence strategy. The cost of advertising and developing new productsmay erode the profits gained by protecting a brand’s market share. Alternativelylaunching an own-brand generic may provide a profit stream that the branded productwould otherwise not receive (Reiffen and Ward, 2007). Understanding the overall profitimplications of the brand defence strategies would provide further evidence of thesuccess of these strategies.

References

Ailawadi, K.L., Pauwels, K. and Steenkamp, J.E.M. (2008), “Private label use and store loyalty”,Journal of Marketing, Vol. 72 No. 6, pp. 19-30.

Aronsson, T., Bergman, M.A. and Rudholm, N. (2001), “The impact of generic drug competitionon brand name market shares – evidence from micro data”, Review of IndustrialOrganization, Vol. 19 No. 4, pp. 425-35.

Barak, A. and Wilson, G. (2003), “Pricing policies to handle patent loss or expiry”, InternationalJournal of Medical Marketing, Vol. 3 No. 3, pp. 245-9.

Baum, C.F. (2006), An Introduction to Modern Econometrics Using Stata, Stata Press, CollegeStation, TX.

Bayus, B.L. and Putsis, W.P. (1999), “Product proliferation: an empirical analysis of product linedeterminants and market outcomes”, Marketing Science, Vol. 18 No. 2, pp. 137-53.

Berndt, E.R., Kyle, M.K. and Ling, D.C. (2003), “The long shadow of patent expiry. Generic entryand Rx-to-OTC switches”, Bureau of Economic Research, pp. 229-67.

Carpenter, G.S. and Nakamoto, K. (1989), “Consumer preference formation and pioneeringadvantage”, Journal of Marketing Research, Vol. 26 No. 3, pp. 285-98.

Caves, R.E., Whinston, M.D. and Hurwitz, M.A. (1991), “Patent expiration, entry, and competitionin the US industry”, pp. 1-66, Brookings Papers: Microeconomics, available at: www.brookings.edu/, /media/projects/bpea/1991%20micro/1991_bpeamicro_caves.pdf

Chandon, P. (2004), “Innovative marketing strategies after patent expiry: The case of GSK’santibiotic Clamoxyl in France”, International Journal of Medical Marketing, Vol. 4 No. 1,pp. 65-73.

Strategies usedto defend brands

1211

Ellison, G. and Ellison, S.F. (2007), “Strategic entry deterrence and the behavior ofpharmaceutical incumbents prior to patent expiration”, working paper no. 13069,National Bureau of Economic Research, Cambridge, MA, April.

Frank, R.G. and Salkever, D.S. (1992), “Pricing, patent loss and the market for pharmaceuticals”,Southern Economic Journal, Vol. 59 No. 2, pp. 165-79.

Gatignon, H., Robertson, T.S. and Fein, A.J. (1997), “Incumbent defense strategies against newproduct entry”, International Journal of Research in Marketing, Vol. 14 No. 2, pp. 163-76.

Grabowski, H.G. and Vernon, J.M. (1992), “Brand loyalty, entry, and price competition inpharmaceuticals after the 1984 Drug Act”, Journal of Law and Economics, Vol. 35 No. 2,pp. 331-50.

Hauser, J., Tellis, G.J. and Griffin, A. (2006), “Research on innovation: a review and agenda formarketing science”, Marketing Science, Vol. 25 No. 6, pp. 687-717.

Hausman, J.A. and Taylor, W.E. (1981), “Panel data and unobservable individual effects”,Econometrica, Vol. 29 No. 6, pp. 1377-98.

Hollis, A. (2002), “The importance of being first: evidence from Canadian genericpharmaceuticals”, Health Economics, Vol. 11 No. 8, pp. 723-34.

Hollis, A. (2003), “The anti-competitive effects of brand-controlled pseudo-generics in theCanadian pharmaceutical market”, Canadian Public Policy – Analyse de Politiques, Vol. 29No. 1, pp. 21-32.

Huskamp, H.A., Donohue, J.M., Koss, C., Berndt, E.R. and Frank, R.G. (2008), “Generic entry,reformulations and promotion of SSRIs in the US”, PharmacoEconomics, Vol. 26 No. 7,pp. 603-16.

Iizuka, T. (2004), “What explains the use of direct-to-consumer advertising of prescriptiondrugs?”, Journal of Industrial Economics, Vol. 52 No. 3, pp. 349-79.

Kalyanaram, G. (2008), “The order of entry effect in prescription (Rx) and over-the-counter (OTC)pharmaceutical drugs”, International Journal of Pharmaceutical and HealthcareMarketing, Vol. 2 No. 1, pp. 35-46.

Kalyanaram, G. and Wittink, D.R. (1994), “Heterogeneity in entry effects between nondurableconsumer product categories”, International Journal of Research in Marketing, Vol. 11No. 3, pp. 219-31.

Kamien, M.I. and Zang, I. (1999), “Virtual patent extension by cannibalization”, SouthernEconomic Journal, Vol. 66 No. 1, pp. 117-31.

Kapferer, J.N. (2008), The New Strategic Brand Management, Kogan Page, London.

Karakaya, F. and Stahl, M.J. (1989), “Barriers to entry and market entry decisions in consumerand industrial goods markets”, Journal of Marketing, Vol. 53 No. 2, pp. 80-91.

Kerin, R.A., Kalyanaram, G. and Howard, D.J. (1996), “Product hierarchy and brand strategyinfluences on the order of entry effect for consumer packaged goods”, Journal of ProductInnovation Management, Vol. 13 No. 1, pp. 21-34.

Kong, Y. and Seldon, J.R. (2004), “Pseudo-generic products and barriers to entry inPharmaceutical markets”, Review of Industrial Organization, Vol. 25 No. 1, pp. 71-86.

Kuester, S., Homburg, C. and Robertson, T.S. (1999), “Retaliatory behavior to new product entry”,Journal of Marketing, Vol. 63 No. 4, pp. 90-106.

Mehta, S.C. and Mehta, S.S. (1997), “Strategic options for brand name prescription drugs whenpatents expire”, Health Marketing Quarterly, Vol. 14 No. 3, pp. 107-14.

EJM46,9

1212

Regan, T.L. (2008), “Generic entry, price competition, and market segmentation in theprescription drug market”, International Journal of Industrial Organizations, Vol. 26 No. 4,pp. 930-48.

Reiffen, D. and Ward, M.R. (2005), “Generic drug industry dynamics”, The Review of Economicsand Statistics, Vol. 87 No. 1, pp. 37-49.

Reiffen, D. and Ward, M.R. (2007), “Brand generics as a strategy to limit cannibalization ofpharmaceutical markets”, Managerial and Decision Economics, Vol. 28 No. 4, pp. 251-65.

Sansgiry, S.S., Bhosle, M. and Pope, N. (2005), “Consumer perceptions regarding generic drugsubstitution: an exploration study”, Journal of Pharmaceutical Marketing & Management,Vol. 17 No. 1, pp. 77-91.

Scott Morton, F.M. (2000), “Barriers to entry, brand advertising, and generic entry in the USpharmaceutical industry”, International Journal of Industrial Organization, Vol. 18 No. 7,pp. 1085-104.

Sethuraman, R. (2009), “Assessing the external validity of analytical results from national brandand store brand competition models”, Marketing Science, Vol. 28 No. 4, pp. 759-81.

Shankar, V., Carpenter, G.S. and Krishnamurthi, L. (1998), “Late mover advantage: how innovativelate entrants outsell pioneers”, Journal of Marketing Research, Vol. 35 No. 1, pp. 54-70.

Steiner, R.L. (2004), “The nature and benefits of national brand/private label competition”,Review of Industrial Organization, Vol. 24 No. 2, pp. 105-27.

Sudhir, K. and Talukdar, D. (2004), “Does store brand patronage improve store patronage?”,Review of Industrial Organization, Vol. 24 No. 2, pp. 143-60.

Sullivan, D.L., Birdwell, S.W. and Kucukarslan, S.N. (1994), “An assessment of consumerpurchasing behaviour for private-label vs brand-name over-the-counter products in chainpharmacies”, Journal of Pharmaceutical Marketing &Management, Vol. 8 No. 1, pp. 85-108.

Sutherland, M. and Sylvester, A.K. (2000), Advertising and the Mind of the Consumer, Allen& Unwin, Sydney.

Thomas, L.A. (1999), “Incumbent firms’ response to entry: price, advertising and new productintroduction”, International Journal of Industrial Organization, Vol. 17 No. 4, pp. 527-55.

Vakratas, D., Rao, R.C. and Kalyanaram, G. (2003), “An empirical analysis of follower entrytiming decisions”, Marketing Letters, Vol. 14 No. 3, pp. 203-16.

Walker, O.C., Boyd, H.W., Mullins, J. and Larreche, J. (2003), Marketing Strategy.A Decision-focused Approach, McGraw-Hill Irwin, New York, NY.

Wiggins, S.N. and Maness, R. (2004), “Price competition in pharmaceuticals: the case ofanti-infectives”, Economic Inquiry, Vol. 42 No. 2, pp. 247-63.

Further reading

Bowman, D. and Gatignon, H. (1996), “Order of entry as a moderator of the effect of themarketing mix on market share”, Marketing Science, Vol. 15 No. 3, pp. 222-42.

Sullivan, M.W. (1992), “Brand extensions: when to use them”, Management Science, Vol. 38 No. 6,pp. 793-805.

About the authorsDean C.H. Wilkie is a PhD Student at the University of Sydney. His research focuses on strategicmarketing with an emphasis on pharmaceutical brand management. His qualifications include aBCom and MCom as well as ten years’ professional experience in brand management. Dean C.H.Wilkie is the corresponding author and can be contacted at: d.wilkie@econ.usyd.edu.au

Strategies usedto defend brands

1213

Lester W. Johnson is the Professor of Management (Marketing) at the Melbourne BusinessSchool, The University of Melbourne. He has a BA from the University of New Hampshire and anMA and PhD from the University of Connecticut. Dr Johnson’s research focuses on customersatisfaction and consumer behaviour in services. He has published in Journal of the Academy ofMarketing Science, Journal of Retailing, International Journal of Research in Marketing, Journalof Service Research, Journal of Advertising Research, European Journal of Marketing, Journal ofInternational Marketing, and numerous other marketing and business journals as well as threebooks. He is currently the Editor of Australasian Marketing Journal and an Associate Editor ofJournal of Service Management. He is a Fellow of both the Australia and New Zealand MarketingAcademy and the Australian Market and Social Research Society.

Dr Lesley White is the Professor of Pharmacy Management at the University of Sydney,where she is coordinating the research and teaching in this new field. Her interests focus onstrategic marketing and services marketing, particularly services quality, together with smallbusiness and professional services marketing. Her qualifications include a BPharm, MCom andMEd. Her PhD is from Sydney University with a thesis addressing decision making in aprofessional services context.

EJM46,9

1214

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints